Chatham House

A primary measure of success for the first Dialogue would be the strength of its contribution to governance interoperability, by reducing fragmentation and duplication across regional and international AI governance instruments. The Dialogue convenors are well-positioned to advance this in two ways. First, the sessions themselves should be designed to build shared language and exchange information among participants, creating the conditions for improved coherence across jurisdictions and institutions. Second, the Dialogue should establish an interoperability resource, such as a public-facing, live document that maps how Dialogue outputs (recommendations, narratives, tools, non-binding commitments) are being channelled into or complementary with (existing and emerging) international approaches. (The proposed Expert Working Group on AI Governance Interoperability can support this in advance of the Dialogue.) This would serve a dual function: helping the process be as clear as possible about its contributions to other global governance instruments, and equipping regulators and policymakers with the information needed to align domestic frameworks with international approaches. This submission notes that many experts in the broader global technology governance space fear over-crowding and duplication of nascent instruments. More effective AI governance is hinged on interoperability, and the Dialogue has a strategic opportunity to demonstrate global collaboration and complementarity in real-time.

Interoperability of governance approaches;

Chatham House's research and convening on global technology governance has focused on how strategic interoperability, alignment and coherence between national and regional regulatory approaches may better mitigate shared (i.e. cross-border, cross-sector) risks and advance global governance outcomes. As covered in Q8, this submission encourages the Dialogue to proactively encourage and platform interoperability as a concrete post-Dialogue outcome.

-

-

The Dialogue can both enable interoperability as a conduit to improved international cooperation (see Q8) and platform different approaches that may enable international cooperation (see Q18).

-

-

Recent Chatham House convening brought together technologists and economists to identify lessons from financial governance for digital governance. This convening demonstrated the value of bringing together expert communities (including from different sectors) with different approaches to governance, risk, uncertainty and evidence. Economists and technologists, for example, frequently disagreed about how to diagnose technological futures, the severity of the challenges (to be) faced and therefore, what solutions and governance are feasible. Rather than smoothing over those tensions, holding space for this range of sectoral and academic bases produces better conversations, learnings and, ultimately, better governance: at a national and global level. https://www.chathamhouse.org/events/all/standard-event/bridging-two-worlds-economic-and-technology-governance This submission encourages global convenors of current and future Dialogues to do agenda-setting at disciplinary intersections (e.g. with finance/economics) and - during the main and side events - draw from a diverse pool of expertise, de-siloising separate interest-based groups. The friction is often where insights lie and governance gaps are filled. [Connection to Q15] Additionally, this submission urges the upcoming and future Dialogues to proactively enable representation from marginalised communities, including racial minorities, gender-diverse and LGBTQIA+ individuals, people with disabilities, and indigenous groups. Human rights organisations and organisations dedicated to representing these communities are best-placed to elaborate on specific recommendations on formats/structures for enabling diverse participation.

-

Recent Chatham House research examined historical case studies - such as the Global Financial Crash (2007/8) and the UK response to the WannaCry ransomware attack (2017) - to generate lessons for crisis- and disruption-proof global AI governance. A key lesson was the importance of leveraging pre-existing institutions, networks of technical expertise and risk-monitoring infrastructure, rather than creating wholly new mechanisms. Drawing from these research findings, this submission urges the Dialogue to consider the following general recommendations that could enable AI governance efforts. These suggestions are focuses on improving disruption- and crisis-preparedness: 1. Develop pre-agreed 'red lines' detailing unacceptable AI risks and generate broad-based consensus. 2. Facilitate diplomatic back-channels and expert-to-expert networks, for example, between industry leaders, UN officials, and state representatives. 3. Leverage any future Global Fund on AI to enable safety-based technical intervention tools, particularly for the Global Majority, such as evaluation infrastructure and supervisory monitoring. 4. Utilise the Global Dialogue and future meetings to prepare shared messages about AI - particularly regarding its potential societal disruption or future cases of malicious- or mis-use with global implications - to generate political buy-in for rapid response, mitigating mis- and disinformation, and misattribution. 5. Contribute to infrastructure for improved institutional information-sharing between governments and the private sector to prevent information asymmetries from hindering crisis response. https://www.chathamhouse.org/2026/03/breaking-deadlock-ai-governance